Introduction for Embedding Infobright for OEMs
- 1. Embedding Infobright Enterprise Edition for Competitive Advantage Confidential – Do Not Distribute 1
- 2. Agenda & Housekeeping Agenda: – The OEM Challenge – Infobright Enterprise Edition for OEMs – OEM Customer Examples – Q&A Housekeeping – Submit questions through the Q&A window – Recording will be available within 24 hours Confidential – Do Not Distribute 2 Michael Hackney, Head of Product Development Jeff Kibler, Director Field Services & Support
- 3. Who is Infobright Global provider of database analytics platforms to over 450 OEM and direct customers in the telecom, digital media and marketing, financial services and solution provider markets.
- 4. As data volumes increase, companies are looking to find more meaningful value in their data.
- 5. Driving Value Out of Data Network Network optimization Troubleshooting Capacity Planning Customer Assurance Fraud Detection CDRs Customer Behavior Analysis Marketing Campaigns/Services Analysis Optimize Network Capacity Fraud Detection Compliance and Audit Advertising Click Through Analytics Engagement Analytics Device Analytics Customer Behavior Analysis Confidential – Do Not Distribute 5
- 6. Your Customer Demands Store more data Deliver answers almost as fast as the data comes in Reduce operational and capital expense Confidential – Do Not Distribute 6
- 7. Solution providers face new challenges as application architectures buckle under the speed and volume of data being generated. Confidential – Do Not Distribute 7
- 8. Solution Provider Battle Shortening time to market in intense competitive environment Scalability issues hindering performance Maintaining margins while delivering increasingly complex services Confidential – Do Not Distribute 8
- 9. Options for Meeting the Challenge Stick with what you have • Pros: -Familiar technology -No integration effort • Cons: - CAPEX - OPEX - Customer satisfaction Rip & Replace: Open Source • Pros: - Lots of choice - No royalty fees • Cons: - OS license restrictions - Time to market - 3rd party support cost Rip & Replace: Commercial • Pros: - Address shortcomings - Supported • Cons: - Cost - Developer learning curve - Proprietary Technology Confidential – Do Not Distribute 9
- 10. Leading technology and solution providers embed Infobright’s analytic database platform to deliver customers’ data management and analytics requirements. Confidential – Do Not Distribute 10
- 11. Infobright Powers Big Data Confidential – Do Not Distribute 11
- 12. How we do it Confidential – Do Not Distribute 12
- 13. Column vs. Row Row Oriented All the columns are needed Transactional processing is required Column Oriented Only relevant columns are needed Reports are aggregates (sum, count, average, etc.)
- 14. The Knowledge Grid Architecture Confidential – Do Not Distribute 14
- 15. Data Loading Process: Compression & Knowledge Grid … … … 64K 64K 64K 64K Data packs compressed On-Disk storage In Memory Knowledge Grid A B C
- 16. The Knowledge Grid: At Work Knowledge Nodes answer the query directly, or Identify only required Data Packs, minimizing decompression, and Predict required data in advance based on workload
- 17. Faster Time to Market: Architectural Flexibility INFOBRIGHT & MYSQL INFOBRIGHT & POSTGRES Confidential – Do Not Distribute 17
- 18. Increased Solution Value Load speeds: – Concurrent loading into single or multiple tables – 2TB+ per hour Query performance – Ad hoc queries that may take hours with other databases run in minutes; – Queries that take minutes with other databases run in seconds Scale – 150TB+ Confidential – Do Not Distribute 18 Knowledge Grid Compressed Data
- 19. Reduced Cost of Goods Sold Reduction in CAPEX – Minimal hardware cost reduction from compression and single server Reduced administrative overhead – No data partitioning, no indexes, no projections, no manual tuning Licensing model – Flexible to meet OEM business models Confidential – Do Not Distribute 19 Original Data 10 TB Compressed Data 500 GB Average compression 20:1
- 20. “Infobright provides real-time data availability and allows users to quickly drill down for ad-hoc analysis and reporting to ensure the highest degree of security for their critical network infrastructure.” Patrick Sweeney, VP, Product Management, Dell SonicWALL Confidential – Do Not Distribute 20
- 21. Customer Example: JDS Uniphase Requirements Low Admin: Do not want to force customers to require DBA’s to keep solution running Load Speeds: Ingestion rates continue to increase, placing heavy burden on solutions High Compression: Want to keep longer histories in less space Lower TCO: Resulting in better value for customers, better margins for providers Results Stripped Away “DBA” tax requirement required by previous versions Ingesting over 1TB/Hour, with significant headroom beyond that Over 3X the retention period and a 5X simultaneous reduction in storage requirement Lower TCO for users, higher margins for JDSU Little to No Admin Fast Load Speeds 20:1+ Compression Exceptional Ad Hoc Query Performance Very Low TCO 21
- 22. Customer Example: Polystar Requirements Query Performance: Ad-hoc queries were often not returning Load Speeds: Slow and cumbersome as volume approached 1 billion records/min High Compression: Different customers need data for different historical periods Lower TCO: Maintain margins while adding additional value to customers Results Queries returned in seconds as a result of Knowledge Grid architecture Data uploaded in near real time allowing Polystar to write xDRs 4x faster Extended data retention enabling customers to 90-180 days of data Lower TCO for users, higher margins for Polystar with cost effective hardware configurations Exceptional Ad Hoc Query Performance Fast Load Speeds 20:1+ Compression Very Low TCO 22
- 23. Built for Solution Providers Flexible pricing model aligned to GTM – Per customer, per server, SaaS, etc. Support – Beta program – 24x7 service level agreements Training – Minimal training required – Provided onsite or remote Confidential – Do Not Distribute 23
- 24. Infobright Delivers Solution Value Fast Load Query Performance Scale up quickly Time to Market Flexible architecture Low learning curve Ease of implementation Reduced CoGs Industry leading compression Lower hardware cost Lower DBA overhead Confidential – Do Not Distribute 24
- 25. Thank you Questions? Confidential – Do Not Distribute 25
Editor's Notes
- #2: Good morning, afternoon and evening. My name is Nikki Gore and I’m the Vice President of Marketing here at Infobright and I would like to welcome you to our webinar Embedding Infobright Enterprise Edition for Competitive Advantage
- #4: For those of you who may not be that familiar with Infobright, we have been around for about nine years providing our analytic database platform to over 450 OEM and direct customers. We are headquartered in North America but we have sales, development and partner offices around the globe and as you can see from this small representation of logos, some market leading solution providers in the telecom, network, security, ad tech and financial service space OEMing our software.
- #5: One of the big challenges facing basically every company on the planet is that as data volumes increase, there is an urgent, or more like critical requirement for them to find meaning in their data so that they can be more agile and make better business decisions
- #6: You can only do this through having the ability to do real analytics 9as opposed to just straight green line reporting) because analytics drives insights; insights lead to greater understanding of customers and markets; that understanding yields innovative products, better customer targeting, improved pricing, and superior growth in both revenue and profits. There are a multitude of solution areas where being able to quickly load and run analytics against data will drive tremendous value for customers.
- #7: As customers are looking to do more analytics on their increasing volumes of data, they also want to do these analytics over larger periods of time to look for trends, etc. This requires solutions to be able to store more data And as quickly as the data is coming in, customers want to be able to start querying but they don’t want to have to make significant investments (if any at all) in additional resources to be able to manage this new requirement for their business.
- #8: What we are seeing with a lot of solution providers, particularly those who have already been delivering some sort of data management and analytics capabilities within their solutions is that their products are breaking under the load of not just the volume but the speed of the data being generated. For those who are just now introducing these capabilities into their solutions, they are faced with a slightly different angle in that they need to set their architectures and infrastructures up and do some predictions on how they will be able to scale.
- #9: So this leads to really the classic product management battles. We are a solution provider so we are very familiar with these battles.
- #10: Let’s explore the options for solution providers when considering improving or introducing analytics capabilities into their solutions. I think under more general capabilities, some companies would look at whether they could build something themselves. In this case, that doesn’t really apply. I mean who wants to build their own database?
- #13: Very Fast Ingestion Ultra Tight Compression Strong Ad Hoc Query Performance Tantamount to indexing everything Minimal Hardware Near Zero Administration No DBA required to establish or maintain the environment Very Low TCO
- #14: Each method has its benefits depending on your use case. Row oriented databases are better suited for transactional environments, such as a call center where a customer's entire record is required when their profile is retrieved. Column oriented databases are better suited for analytics, where only portions of each record are required. By grouping the data together like this, the database only needs to retrieve columns that are relevant to the query, greatly reducing the overall time and I/O needed. And by contrast, returning a specific 'record' would require retrieving information from each column store. Plus, through our deep compression and intelligence, Infobright reduces the I/O as much as possible to give you as much of the resources back as possible. Infobright is a column oriented database and built for high speed and complex analytical queries that ask questions about the data such as trends and aggregates, rather than questions that retrieve records from the data.
- #15: Foundation in rough set math applied to granular computing concepts Assumes a specific data structure Limited number of tables Extremely large record counts Once a record is written, it is seldom if ever updated Data Packs are assembled into a metadata layer or Knowledge Grid Queries are run iteratively, first as a “rough evaluation” over the Knowledge Grid, then an “exact evaluation” over the compressed data as needed
- #17: Need to say that Global knowledge is contained within every data pack The Knowledge Grid is a summary of statistical and aggregate information collected about each table as the data is loaded. Its information about the data. For each column and each Data Pack within that column, the Knowledge Grid information is collected automatically and different types of Knowledge Nodes are built with no configuration or setup required in advance. For example, three of the Knowledge Nodes, called Data Pack Nodes, Numerical Histograms, and Character Maps, are built for each Data Pack during the load; This eliminates the need for indexes Other, dynamic Knowledge Nodes, are built when more complex queries are run. Some eliminate the need for keys. Others eliminate complex aggregate intermediate results. Because they contain summary information about the data within the table, the Knowledge Nodes are used as the first step in resolving queries quickly and efficiently by answering the query directly, or by identifying only relevant Data Packs within a table and minimizing decompression.
- #18: Infobright uses widely known APIs for connections to programming language and tools You can use the MySQL Connectors and Postgres Connectors We have the same structural query language We can easily be a drop in solution to start and then optimize in future revisions Most OEMs have some transaction piece – insert the data and update the data quickly